Estrogen activates endothelial exocytosis

Abstract

Estrogen therapy is used to treat patients with post-menopausal symptoms, such as hot flashes and dyspareunia. Estrogen therapy also decreases the risk of fractures from osteoporosis in post-menopausal women. However, estrogen increases the risk of venous thromboembolic events, such as pulmonary embolism, but the pathways through which estrogen increase the risk of thromboembolism is unknown. Here, we show that estrogen elicits endothelial exocytosis, the key step in vascular thrombosis and inflammation. Exogenous 17β-estradiol (E2) stimulated endothelial exocytosis of Weibel-Palade bodies (WPBs), releasing von Willebrand factor (vWF) and interleukin-8 (IL-8). Conversely, the estrogen antagonist ICI-182,780 interfered with E2-induced endothelial exocytosis. The ERα agonist propyl pyrazole triol (PPT) but not the ERβ agonist diarylpropionitrile (DPN) induced vWF release, while ERα silencing counteracted vWF release by E2, suggesting that ERα mediates this effect. Exocytosis triggered by E2 occurred rapidly within 15 min and was not inhibited by either actinomycin D or cycloheximide. On the contrary, it was inhibited by the pre-treatment of U0126 or SB203580, an ERK or a p38 inhibitor, respectively, suggesting that E2-induced endothelial exocytosis is non-genomically mediated by the MAP kinase pathway. Finally, E2 treatment enhanced platelet adhesion to endothelial cells ex vivo, which was interfered with the pre-treatment of ICI-182,780 or U0126. Taken together, our data show that estrogen activates endothelial exocytosis non-genomically through the ERα-MAP kinase pathway. Our data suggest that adverse cardiovascular effects such as vascular inflammation and thrombosis should be considered in patients before menopausal hormone treatment.

Keywords: Endothelial cells; Estrogen; Exocytosis; Platelet; Thromboembolism.

Figure 1.

Estrogen activates endothelial exocytosis. HUVECs were treated with 17β-estradiol (E2) for 1 hr and the release of VWF or IL-8 was measured by ELISA. (A) Dose response for vWF release. Increasing concentrations of E2 cause a dose dependent increase in endothelial release of vWF (N=4). (B) Dose response for IL-8. Increasing concentrations of E2 cause an increase in endothelial release of IL-8 (N=3). All data from A-B are presented as mean ± SEM (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Figure 2.

Estrogen activates endothelial exocytosis through the ERα receptor. HUVECs were treated with E2, inhibitors or agonists of estrogen signaling and the release of vWF was measured by ELISA. (A-B) The estrogen receptor inhibitor ICI 182,780 (ICI, 100 nM) was pre-treated 1 hr before E2 (100 nM) treatment (A) or histamine treatment (B) and vWF release was measured. (C) 100 nM of mifepristone (M) or spironolactone (S) was pre-treated 1 hr before E2 treatment, and vWF release was measured. (D) HUVECs were treated with the ERα agonist, PPT, or the ERβ agonist, DPN for 1 hr, and vWF release was measured. (E-F) ERα was silenced 2 days before E2 treatment (E), and vWF release was measured (F). All data from A-F were repeated three times and are presented as mean ± SEM (*, P < 0.05; **, P < 0.01; ***, P < 0.001; n.s., not significant).

Figure 3.

Estrogen activates endothelial exocytosis through rapid, non-genomic pathways. HUVECs were treated with E2 and inhibitors of transcription, translation or MAPK signaling, and the release of VWF was measured by ELISA. (A) Time course study: E2 (100 nM) causes endothelial cells to release VWF over time (N=4). (B) Actinomycin D (2.5 μM) was pre-treated for 2 hr before E2 treatment (N=3). (C) Cyclohexamide (1 μM) was pre-treated for 6 hr before E2 treatment (N=3). (D) U0126 or SB203580 (10 μM) was pre-treated for 1 hr before E2 treatment (N=4). All data from A-D are presented as mean ± SEM (*, P < 0.05; **, P < 0.01; ***, P < 0.001; n.s., not significant).

Fig 4.

Estrogen treatment enhances in vitro platelet adhesion to endothelial cells. (A-B) HUVECs were pre-treated with ICI-182,780 for 1 hr and then treated with E2 or histamine for 1 hr. Then, BCEF-AM loaded platelets were co-cultured with HUVECs at 4 °C for 15 min and imaged with a digital fluorescent camera. Representative figure (A) and bar graph (B) of the number of platelets adhered to HUVECs were shown. (C-D) HUVECs were pre-treated with U0126 for 1 hr before E2 treatment. BCEF-AM loaded platelets were co-cultured with HUVECs at 4 °C for 15 min and imaged with a digital fluorescent camera. Representative figure (C) and bar graph (D) of the number of platelets adhered to HUVECs were shown. All data from A-D were repeated four times and are presented as mean ± SEM (*, P < 0.05; **, P < 0.01; ***, P < 0.001). All scale bars, 10 μm.